Every year in the U.S., more than 1.5 million rear-end car accidents occur, injuring over 500,000 people. In order to identify the visual and perceptual functions related to these accidents, we developed a portable test of looming and added more realistic tests of the same construct. These tasks in the battery require the detection, judgment, and manual control of (a) motion in depth, (b) transverse movement in depth, (c) self-motion, and (d) light signals. In Phase I through a graded progression in fidelity from single tasks to a complex driving simulation, we were able to demonstrate that it is feasible to (a) identify the critical performance elements involved in this aspect of the driving task, (b) identify those task elements that are most affected by age and gender, and (c) select those perceptual-motor tasks that appear most promising for modification through training. We found that age effects showed negative correlations between looming performance and more realistic renditions of a looming task including a desk-top driving simulator. in addition to the potential for scientific investigations, the proposed apparatus has application for driver testing, licensing, and perhaps, rehabilitative training, and it may help to converge and extend previously unrelated findings in visual science and aging. PROPOSED COMMERCIAL APPLICATION: We believe that our laboratory looming test and the battery old more realistic tasks will find a wide market among those interested in the performance of the elderly related to driving research, driver testing, and driver training. In particular, we expect the device to complement existing driver assessment and training devices. There are other areas of application: information and education of elderly drivers; training (if loom is trainable) or alternate driving strategies for loom dependent tasks; selection for certain occupations (if loom is not trainable).